You may have heard about KickStarter and other crowd funding sites. These sites help people raise money for new ventures. The idea is that if a lot of people invest a small amount of money, a project can get funded that may not be up to the rigorous standards of venture capital investors (VCs). VCs normally want a 70% return. But with sites like KickStarter or GoFundMe, the investor may only want a T-Shirt from the project or one of the first runs of a new product. Often, this is a way to see a cool technology realised – and a chance to get in on the ground floor.
But there is a darker side or crowd sourcing – graphic design projects. Before we go down the dark path, let’s take a look at a good business example.
The good part of using a crowd, is that often the crowd gives you better information than a small group or one person would, very Lean approach some would say. For example, guessing how many beans there are in a jar. Often you see a jelly-bean jar content at fairs. Each person buys a ticket and tries to guess the number of beans in the jar. The winner is the person who comes the closest. And normally that would be the end of it.
One statistician – Michael Mauboussin – discovered, that any person trying to guess the beans in a jar is way off – no surprise. But the MORE people who entered the contest, the closer the average of the guesses was to the correct total. This is quite amazing. If only ten people guess, the average may be off by 30%. But ask 100 people to guess, and the average is only off by 10%. Get 1,000 people, and the discrepancy may only be off by one bean.
So it seems there is wisdom in crowds.
And so business started to use this idea. Threadless is a company that prints T-Shirts. But their twist, is that designers submit design ideas that are viewed online. Only if enough people say the’d buy the shirt is it printed – and the designer paid.
Seems fair enough. Both parties win. If the market (the people who view and rate the design) don’t like it, no shirts are printed and the designer gets nothing. This is the basic risk any designer takes – no market acceptance – no money. But now the T-Shirt company is spared the expense and risk of printing shirts without orders.
If the designer has success (the market loves the design and places orders) the T-Shirt company prints the shirts knowing the sales are there already – no risk. With orders, the print shop is more than happy to pay the designer. In fact, they’ll pay $2,000 if a design results in orders.
All seems fair.
Now enter 99designs. A company that uses crowd sourcing to fulfil demand for graphic design projects (T-Shirt designs, Logos, Web banners etc.). The idea is simple. A company pays 99designs to run a contest. The contest will pay the winning designer a prize if the client likes their design. Sounds simple.
But in a bad economy, each client that wants a logo may get 30 designs. Lots of designers are desperate for work. With a prize of $300, it seems like good pay for a designer who works on the logo for three hours.
This may seem like crowd sourcing, but it’s not. It’s the dark side of the internet.
In an average design contest, the client gets 30 designs from 30 designers. In theory, more is better. But this model does not actually produce better results. It’s not like the beans, where the average is closer to the real number. More designers just produce more … but not better. There is no cumulative effect.
Looking at the design contests, I noticed that most of the designs are very similar. The more the client specifies design colors, and fonts, the closer all the designs are.
More importantly, the numbers don’t work. Taken as a group, if 30 designers work three hours each on a logo design, you have 90 hours of work, for $300. That’s just $10/3 hour or $3.33 per hour. Economically, while one person wins, the others all loose. But the cumulative effect would be under $4 per hour for the effort.
99design published a list of their top designers. Shockingly, they published how many contests each designer entered, and how many they won. The top designers (based on the number of contests they won) showed a staggering loss. On average, the top designers, only won 10% of the contests they entered. 99Designs has since pulled this advertisement. It was meant to entice other designers to want to compete in design contests. But anyone with basic math skills could divide the amount these designers won, over the number of contests they entered to see how little they actually earned per hour. At $300 per win, and a win rate of 10%, they averaged $30 per contest entered. Not bad pay of one hour of work. Easily it takes three hours to read the client design brief, do the design work, and make three variations in hopes of winning the contest. If so, the designers are earning $10 per hour.
This is the dark side of crowd sourcing, let’s call it slave sourcing. While it seems more people add to the quality, often one designer, paid $300 would likely do a better job, and spend more time on the client’s logo.
For good or bad, the internet makes crowd sourcing easy. And companies are trying to find ways to entice groups to lend the power of the crowd to their project.